• Quikscat measurements of the wind field over the Bering and Chukchi Seas

      Mull, Jeremy M.; Johnson, Mark; Weingartner, Tom; Simmons, Harper (2008-12)
      The purpose of this study is to investigate the dynamic wind field and resulting ocean circulation patterns in the Bering and Chukchi Seas. This region forms an important link in global ocean circulation as Bering Strait is a major conduit for water flowing into the Arctic Ocean. The Arctic has been identified as an area sensitive to climate change; thus it is vital to understand how water and energy flow through this region. We first quantify the differences between the winds measured in this region by the Quik Scatterometer (QuikSCAT) and those modeled by the National Centers for Environmental Prediction (NCEP). Although the data sets are well correlated, we find significant discrepancies between these data sets and use linear regressions to correct the NCEP data. The magnitudes of the NCEP wind components are greater than the magnitudes of the QuikSCAT wind components. This creates directional differences between the two data sets at low wind speeds and NCEP speeds that are greater than QuikSCAT speeds at high wind speeds. We next challenge the assumption that the wind field is spatially uniform over the Bering and Chukchi shelves. We produce mean monthly maps of the wind field, surface Ekman transport, and wind variance based upon the 12-hourly QuikSCAT data from July 1999 – May 2007. These maps reveal that the winds are spatially and temporally dynamic in this region. There are several areas and times in which surface Ekman transport is onshore or offshore near the coasts and may engender coastal downwelling and upwelling, respectively. There are also several instances when surface Ekman convergence and divergence may lead to Ekman pumping and suction. We use the entire NCEP record (January 1948 – May 2007) to examine patterns of surface Ekman transport across the shelf break. There was a significant increase in the amount of onshelf surface Ekman transport that coincided with the regime shift that occurred in the Bering Sea in the mid-1970s. We attempt to correlate the time series of surface Ekman cross-shelf transport with several climate indices but find only very weak correlations. The annual surface Ekman freshwater fluxes across the shelf break are iii calculated and found to be very small compared to the total annual freshwater fluxes calculated by Aagaard et al. (2006) and Kinney et al. (2008). To resolve the dominant modes of wind variability we compute hourly, monthly, and annual Complex Empirical Orthogonal Functions (CEOFs) with the QuikSCAT and NCEP data sets. The first modes in each analysis account for more than 60% of the variance. Different aspects of the mode amplitude time series are cross-correlated with climate and indices to produce small but significant correlation coefficients. Finally we calculate Ekman pumping and suction at four locations in the Bering Sea during the spring and summer months of seven years (2000 – 2006). We identify regions and times when Ekman pumping and suction were particularly strong, and perform several runs of a one-dimensional Price-Weller-Pinkel (PWP) vertical mixing model with the QuikSCAT winds, the QuikSCAT winds and a wind-stress-curl term, and the NCEP winds. The results suggest that Ekman suction might facilitate subsequent vertical mixing while Ekman pumping might inhibit subsequent vertical mixing when the winds are generally weak and wind-stress-curl is moderate or strong. The temporal resolution of the QuikSCAT data set is too low to resolve inertial motions at high latitudes. The NCEP data set has higher temporal resolution and is adequate for running the model within this region. We propose interpolating the hourly NWS data collected at St. Paul Island (station PASN) to the QuikSCAT grid using the complex amplitudes and phases from the complex cross-correlations between the two data sets to produce a data set of high temporal and spatial resolution. This would enable researchers to accurately resolve inertial motions and compute wind-stress-curl.